The proposed research integrates two lines of research on childhood aggression and victimization: one focusing on forms and functions of (individual differences in) aggression and victimization, and the second focusing on aggressor-victim relationships. The proposed research integrates these two lines by studying various forms (overt and relational) and functions (proactive and reactive) of aggressor-victim relationships. The specific aims are to (1) evaluate the psychometric properties of a newly-created instrument for identifying various types of aggressor-victim relationships;(2) evaluate the magnitude of overlap among these various forms of aggressor-victim relationships;and (3) identify concurrent correlates of involvement in these various aggressor-victim relationships. Data will be collected from approximately 500 sixth through eighth grade boys and girls from schools with primarily Hispanic and non-Hispanic White students (with smaller representations of other ethnic groups). The key instrument is the newly-created Dyadic Forms/Functions of Aggression and Victimization Inventory (DFFAVI) designed to identify overt, relational, proactive, and reactive types of aggressor-victim relationships. Other instruments will be used to evaluate the validity and concurrent correlates of this instrument. These data will be analyzed using multivariate social relations modeling. Given the prevalence of aggression and victimization, as well as associated maladjustment, the proposed research is expected to provide valuable information regarding the relationship contexts in which various forms and functions occur. The approach to studying forms and functions of aggressor-victim relationships is innovative in being the first to integrate these two lines of research, and will advance our understanding of the various types of aggression in the dyadic contexts in which they occur. The proposed research is expected to also serve as valuable groundwork for subsequent longitudinal investigation of these relationships.